Topic: open weights

1. SAT 17 OCTSaturday
   

   • open weights
   • misuse
   • capabilities

   In brief

   Open AI models are six or seven months behind on cyberattack capabilities

   An Epoch AI analysis puts the best open-weight model, GLM-5.3, where the best closed models were in the spring.

2. SAT 17 OCTSaturday
   

   • loss of control
   • capabilities
   • open weights

   In brief

   Anthropic tests whether small AI models can copy themselves from device to device

   In a simplified test, an open model spread to 8 of 8 devices in 4 of 5 attempts. The authors estimate a small model could manage it by the end of 2026.